In a digital wireless communication, an adaptive array antenna (hereinafter referred to as “AAA”) technique is used which controls adaptively directivity by adding a weight to an antenna output of each of a plurality of antenna elements. In the AAA technique, it is possible to suppress interfering signals by adaptively controlling the directivity using the fact that the directions of arrival of signals are mutually different. Therefore, the adaptive array antenna technique is important as a method of canceling interfering signals on the same channel.
A radio base station apparatus that performs the AAA processing will be described. In the radio base station apparatus, signals received via a plurality of antennas are subjected to the predetermined processing such as radio reception processing and then to hard decision. Then, an error signal is generated based on the received signal with no hard decision performed thereon and the received signal (hard decision data) with the hard decision performed thereon, and based on the generated error signal, weights are calculated using an adaptive algorithm such as LMS algorithm and RLS algorithm. The radio base station apparatus multiplies received signals from antennas by respective calculated weights to form directivity, and thereby is capable of suppressing interfering signals and of improving transmission quality.
Generally, in wireless communications, it sometimes happens that the hard decision on a received signal fails due to fading and the like, and that erroneous hard decision data is obtained. A weight calculated based on the erroneous hard decision data has low accuracy, and there is a problem that desired directivity is not obtained based on the erroneous hard decision data. This problem is solved by performing error correcting processing such as Viterbi decoding and interleaving on a received signal to improve the reliability of hard decision data.
However, improving the reliability of hard decision data by the error correcting processing increases the number of processing steps of calculating the hard decision data by 1, resulting in a problem that processing delay occurs and thereby the transmission efficiency deteriorates. Since, in order to improve the error correcting effect in the error correcting processing, it is necessary to increase the constraint length to process increased data, the above-mentioned problem becomes more remarkable as the hard decision data with higher reliability is desired.